Support different number of mismatches per barcode set or 1 number of mismatches to allow the same number of mismatches for all barcode sets.

Issue #5 new
Robert Leach created an issue

USE CASE: WHAT DO YOU WANT TO DO?

Allow the number of mismatches to be supplied as 1 for everything (each barcode allows 1 mismatch), 1 for a set (number of mismatches allowed for each barcode in a single column), or 1 per individual barcode.

STEPS TO REPRODUCE AN ISSUE (OR TRIGGER A NEW FEATURE)

  1. Supply a flag that takes 1 number of mismatches for a set or multiple numbers of mismatches (one for each in a set). [Optionally takes numbers of mismatches per barcode in the barcode file]

CURRENT BEHAVIOR

There's only 1 barcode length that's universally applied to every barcode.

EXPECTED BEHAVIOR

Barcode lengths should be able to be supplied as a single length of every barcode, a length for each barcode set, and/or a length for each barcode in a set.

DEVELOPERS ONLY SECTION

SUGGESTED CHANGE (Pseudocode optional)

Implement similar to the index read flag.

LEVEL OF EFFORT - developers only

major

COMMENTS

It may even be desirable to implement number of mismatches per row of barcodes. It could even be a bound, e.g. a max of 3 mismatches total allowed for a row of 3 barcodes and a max of 2 mismatches allowed in a single barcode. So if the first barcode has 2 mismatches and the second has 1, the third would not be allowed to have any mismatches. This implementation (of a number of mismatches per row of barcodes) would necessitate changing the way mismatches are handled.

Comments (4)

  1. Robert Leach reporter

    If I understand what you mean, the answer is yes. To make sure I understand, I'll clarify. Currently, if you specify 1 mismatch and there are 2 columns of barcodes, on one row, each of the 2 barcodes is allowed a single mismatch.

  2. Robert Leach reporter
    • edited description

    Corrected some wording and added a new consideration for bounded numbers of mismatches allowed (per row and per barcode).

  3. Log in to comment